[unreadable] [unreadable] The goal of this application is to complete a pilot study to determine the feasibility of conducting a large, family-based, genetic-epidemiologic study of breast cancer in Hong Kong, China. In the full-scale study, we plan to obtain pedigree information on all eligible population-based cases (probands) of breast cancer, blood samples and risk information on cases and selected family members, tumor blocks and samples of fresh frozen tissue from probands and tumor blocks from relatives with breast or ovarian cancer. These data will be used to estimate BRCA1 and BRCA2 mutation carriage, to improve models to estimate mutation carriage in different ethnic groups, and to perform case-control studies of breast cancer risk factors, with an emphasis on inherited and candidate genes in selected metabolic pathways, including gene-gene and gene-environment interactions. This application is seeking support for a pilot study to provide information to finalize a study design and to demonstrate the feasibility of each element of the larger study. Components of the pilot study are modeled after the NCI-funded Breast Cancer Family Registry (Breast-CFR). Using these protocols and data instruments, with as few modifications as possible, will allow comparisons of data from the Breast- CFR and this proposed study. The specific aims of this study are to: 1) collaborate with eight hospitals in Hong Kong to prospectively identify and recruit 48 cases (probands) of breast cancer (6 from each hospital) and from each proband obtain fresh frozen tissue of the cancer and surrounding tissue, paraffin-embedded tumor blocks, blood samples, risk factor information, and a pedigree extended to second-degree relatives; 2) recruit 2 relatives of each proband (N=96), obtain blood samples and risk factor information, and tumor blocks and pathology reports for those diagnosed with breast or ovarian cancer; 3) recruit 24 "control probands" (3 per hospital) with no history of breast cancer and 2 family members per proband (N=48) and collect the same information as in the cases families; and 4) estimate response rates to all parts of the study, ability to collect family history information, determine the number of relatives that may be included in the study, determine the prevalence and distribution of risk factors of interest, identify the most efficient study procedures, identify all costs, and evaluate the feasibility of including controls in the large study. [unreadable] [unreadable] [unreadable]